Bariatric surgery reduced the incidence of type 2 diabetes by 78% compared with usual care at 15 years in a prospective, case-matched study of more than 3,000 obese adults.
This significant risk reduction was seen with all types of bariatric surgery and regardless of baseline body mass index. And, it occurred despite the fact that the matching process unexpectedly resulted in the bariatric surgery group having a higher mean body weight and more severe risk factors at baseline than the controls.
The impact of bariatric surgery was even greater, with an 87% risk reduction, for those with impaired fasting glucose at baseline, said Dr. Lena M.S. Carlsson of the Sahlgrenska Academy at the University of Gothenburg, Sweden, and her associates (N. Engl. J. Med. 2012;367:695-704).
"Our data indicate that bariatric surgery has a preventive effect on incident type 2 diabetes, particularly in participants with impaired fasting glucose. In contrast, baseline BMI did not influence the preventive effect of bariatric surgery on type 2 diabetes, implying that anthropometric data are not useful in the selection of candidates for bariatric surgery, whereas data on impaired fasting glucose may be helpful," the authors wrote.
The finding comes from the Swedish Obese Subjects (SOS) trial, which included 1,658 patients who chose to undergo bariatric surgery and 1,771 matched controls. All patients in both groups entered the study with the intention of losing weight. None had diabetes at baseline.
In the bariatric surgery group, the types of procedures were banding in 311, vertical banded gastroplasty in 1,140, and gastric bypass in 207. Patients in the control group received the customary treatment for obesity at their primary health care centers, which in Sweden ranges from advanced lifestyle modification – including recommendations regarding eating behavior, food selection, energy intake, and physical activity – to no treatment. About half (54%) of the controls reported receiving professional guidance in attempts to lose weight.
There were several significant differences between groups at baseline. The bariatric surgery group weighed an average of 6 kg more than did the controls, and had a greater mean BMI (42.4 vs. 40.2 kg/m2). They also had higher mean blood pressures and total cholesterol and triglyceride levels, and were more likely to smoke and to be less active.
After adjustment for follow-up of less than 15 years and for death, the 15-year participation rate was 54%. At 15 years, the bariatric surgery group had lost 31 kg after 1 year, but then regained weight, so the average loss at 10 and 15 years was about 20 kg. The control group never lost or gained more than 3 kg over the entire study period, regardless of whether they had professional help.
During the follow-up, type 2 diabetes developed in 110 of the bariatric surgery patients and in 392 controls, corresponding to incidence rates of 6.8 and 28.4 cases per 1,000 person-years, respectively (P less than .001). The unadjusted hazard ratio was 0.22, which dropped to 0.17 following multivariate adjustments. Aside from treatment group, other strong univariate predictors of diabetes outcome were baseline blood glucose and the presence or absence of impaired fasting glucose, Dr. Carlsson and her associates reported.
In a sensitivity analysis performed to account for the low participation rate, the impact of treatment on the incidence of type 2 diabetes was at least as strong after 2 years and 10 years of follow-up as after 15 years. All types of bariatric surgery were associated with a reduced incidence of diabetes, with no significant differences among them. There were also no differences by receipt of professional weight-loss assistance, or by BMI at baseline, the investigators noted.
A total of 3 patients (0.2%) died within 90 days of surgery, and 245 patients in the surgery group (15%) reported at least one complication. Of those, 46 (2.8%) were serious enough to require a reoperation.
The risk reduction seen among those with impaired fasting glucose was at least twice as large as the risk reduction achieved with lifestyle interventions in large, long-term trials of moderately obese people with prediabetes (Lancet 2006;368:1673-9, Lancet 2009;374:1677-86, and Lancet 2008;371:1783-9), the investigators noted.
The ongoing SOS study is supported by grants from the Swedish Research Council, the Swedish Foundation for Strategic Research to the Sahlgrenska Center for Cardiovascular and Metabolic Research, the Swedish federal government, the VINNOVA-VINNMER program, and the Wenner-Gren Foundations. The SOS study has previously been supported by grants to one of the authors from Hoffmann-La Roche, AstraZeneca, and other companies. Dr. Carlsson reported receiving consulting fees from AstraZeneca and owning stock in Sahltech. Other coinvestigators also had financial disclosures.